A Network Management System (NMS) provides two important services to the Operations Support Systems (OSS) that comprises any known EMS based network. One typical example is the AT&T Competitive Network Initiative (CNI) project support systems complex. The first is to maintain a complete, up-to-date network inventory and status of all installed CNI Element Management Systems (EMS) equipment and their subtending Network Elements (NE). This inventory is updated in real-time via asynchronous notifications from the EMSs or can be updated synchronously through queries invoked using remote procedure calls to the EMS. The second service that NMS provides is to act as a single point of interface through which the CNI OSSs communicate with the EMSs for service related requests. This allows an OSS to request end-to-end EMS based network connections and to receive inventory and change of status related asynchronous notifications as a result of those requests.
Due to the complexity of the interactions between the NMS, other OSSs and the EMSs in normal operation, it has been necessary to do much of the system testing and all of the integration testing for each release in the laboratory test environment using their small, but complete, network of vendor supplied EMSs and OSS platforms. Normally, it is required to execute a complete set of functionality and interoperability testing with all of the latest software releases running on servers identical to those targeted to be installed in the field, but this type of configuration requires many pieces of equipment to test elements, so it is expensive and time consuming with several testing groups competing for an extremely scarce resource. Additionally, constant changes to the network configuration make it impossible to build sets of repeatable tests, such as those needed for regression testing and make interpretation of test results more difficult.
All of the telecommunications providers who are deploying modern EMS based networks face the same problem of being unable to thoroughly test network management systems due to the unavailability of realistic network test setups due to cost, complexity, short delivery schedules, etc.
In order to address these problems, a set of EMS simulators is constructed using data extracted from EMS equipment running the software version intended to be deployed to the field. In effect, a virtual model of a complete EMS test network is built which is capable of faithfully reproducing all of the functionality provided by EMS systems for complete integration testing and capable of being further modified to add any functionality required up to a complete simulation of any number of EMSs including interconnectivity.